Consolidate SAS 9.4 workloads with Intel Xeon processor E7 v3 and Intel SSD technology

APRIL 2015
A PRINCIPLED TECHNOLOGIES TEST REPORT
Commissioned by SAS
CONSOLIDATE SAS® 9.4 WORKLOADS WITH INTEL® XEON® PROCESSOR
E7 V3 FAMILY AND INTEL® SSD DATA CENTER FAMILY
Is an outdated data center holding back your business? It might be time to
modernize and simplify with new systems that can do the work of many older servers.
This can translate to savings in space and operating costs while leaving headroom for
future growth.
For businesses running SAS® Analytics software, this means virtualizing bare-
metal legacy servers onto a newer Intel® Xeon® processor E7 v3 platform, such as the
new Dell PowerEdge R930. In the Principled Technologies data center, we found that a
server powered by Intel Xeon processors E5-2680 v2 with Intel Solid-State Drives (SSD)
DC S3700 Series did the SAS work of nearly four legacy servers, and a newer PowerEdge
R930 powered by Intel Xeon processors E7-8890 v3 did the SAS work of 12 legacy
servers.1
With 12 VMs, the server powered by Intel Xeon processors E7 v3 with Intel SSD
DC P3700 Series for PCIe® delivered nearly 14 times the relative performance of the
legacy server. What’s more, each VM on the server powered by Intel Xeon processors E7
v3 completed the SAS multi-user scenario workload an average of over 25 minutes
sooner, and completed nearly 110 more jobs per hour on average than the bare-metal
1
For more information on the SAS 9.4 workload and jobs, see The SAS workload section on page 3 and Appendix A.

A Principled Technologies test report 2Consolidate SAS 9.4 workloads with Intel Xeon processor E7 v3 and
Intel SSD technology
legacy server did. In addition, some performance capacity remained when running 12
virtual instances simultaneously on the newer server, potentially leaving room for more
virtual SAS 9.4 instances.
SAVING WITH MODERNIZATION
One approach to the modernization of data centers combines consolidating
resources through virtualization with upgrading to new hardware technology. These
improvements can lead to reduced complexities throughout the technology stack, and
potentially reduce operational expenses (OPEX). Specifically, consolidating physical
hardware in the data center typically provides these advantages:
 Reduced power and cooling requirements
 Smaller physical footprints, such as reduced number of racks and
servers, which reduces space-related costs
 Lower management costs due to fewer physical servers to manage
 Fewer resources wasted as a result of underutilized hardware
Newer systems generally support, and will have, substantially more RAM and
storage, sometimes in the form of faster SSDs. These technologies are ideal for in-
memory data processing and analytics, such as SAS 9.4. The additional RAM and high-
performance SSDs work with Intel Xeon processor E5 v2 and E7 v3 technology, and are
designed by Intel to deliver fast performance for analytics workloads while increasing
data center efficiencies. Moving from multiple legacy servers onto newer servers
powered by Intel Xeon processors E7-8890 v3 and Intel SSDs DC P3700 running SAS 9.4,
can help in consolidation efforts, contributing to your modernization initiatives. In our
tests, which we describe further below, we found that one newer four-socket Dell
PowerEdge R930 server powered by Intel Xeon processors E7-8890 v3 could replace 12
legacy servers running SAS workloads (see Figure 1).
Figure 1: Consolidating legacy servers is a key to modernization.

THE RIGHT COMBINATION WITH INTEL + SAS
Deployments of SAS 9.4, an environment designed by SAS for business and
advanced complex data analytics, can benefit from servers powered by the Intel Xeon
processor E7 v3 family as processing, data storage, and analysis requirements increase
over time. The updated Intel technology in four-socket servers can make them well
suited to support large-volume, complex data analytics software such as those from
SAS. The Dell PowerEdge R930, which contains four Intel Xeon processor E7 v3
processors, provides up to 72 cores and 144 threads and supports up to 6TB of DDR4
memory to improve performance and help speed up your SAS workloads. Newer servers
can also support increased RAM and faster SSD storage. In our testing, we took
advantage of these RAM and SSD improvements. The servers powered by Intel Xeon
processors E5-2680 v2 had 256 GB RAM with DC S3700 SATA SSDs and the servers
powered by Intel Xeon processors E7-8890 v3 had 1,024 GB RAM with DC P3700 PCIe
SSDs.
For the modern data center, servers based on the Intel Xeon processor E7 v3
family also offer a number of enhanced reliability, availability, and serviceability (RAS)
and resource management features that build on previous processor releases. With
Intel Run Sure technology, these processors add new RAS features such as eMCA Gen 1,
MCA Recovery – Execution Path, MCA I/O, and PCIe Live Error Recovery to help keep
SAS workloads up and running. For more information on components used in our
testing, see Appendix A. For detailed hardware configuration information, see Appendix
B.
The SAS workload
For our hands-on testing, SAS created a multiuser workload to simulate the
workload of a typical SAS foundation environment consisting of 25 different data
analysis tasks. The test scenario was a self-contained SAS 9.4 workload that simulated 8
to 16 simultaneous users performing computation-focused jobs, as well as I/O-focused
jobs. The data analysis jobs varied from small and lightweight (e.g., simple reporting), to
more complex analytics (e.g., regression), to heavy data manipulations (e.g., joins,
sorts).
For more information on SAS 9.4 and the SAS workload, see Appendix A.
Appendix C shows how we performed our tests. See Appendix D and Appendix E for
configuration files and test scripts.

WHAT WE FOUND
More jobs in less time and more headroom with 12 VMs
Running 12 virtual SAS 9.4 instances on the server powered by Intel Xeon
processors E7-8890 v3 and PCIe SSDs provided a number of benefits. First, we found
that simply virtualizing the SAS workloads increased the amount of jobs that both the
two-socket server powered by the Intel Xeon processors E5-2680 v2 and the four-socket
server powered by the Intel Xeon processors E7-8890 v3 could perform simultaneously.
Simply put, virtualizing meant more work in less physical space, which creates an
opportunity to consolidate legacy two-socket servers in your data center.
Second, the two-socket server powered by Intel Xeon processors E5-2680 v2
provided up to nearly four times the relative performance of the two-socket legacy
server, while the four-socket Intel Xeon processor E7-8890 v3 server provided nearly 14
times the relative performance. This performance calculation uses the average time to
complete the workload to compare the work of each VM. Figure 2 shows the relative
performance for our three solutions.
Figure 2: The relative
performance of each solution.
Keeping those two previous points in mind, the four-socket server powered by
Intel Xeon processors E7-8890 v3 provided nearly 14 times the relative performance of
the legacy server in 14.2 percent less time—completing the SAS workload over 25
minutes faster. The shorter average time to complete the workload was driven by the
four Intel Xeon processors E7-8890 v3 processors and the improved I/O performance
offered by the Intel DC P3700 PCIe SSDs. Improved technology simply meant less time to
complete the work. Figure 3 shows the average time each solution took to complete the
SAS multi-user scenario workload.
Virtualizing meant
more work in less
physical space—up to
12 times the amount
of jobs ran
simultaneously on
one four-socket
server powered by
Intel Xeon processors
E7-8890 v3 as
compared to the
legacy server.
The four-socket
server with Intel
Xeon processors E7-
8890 v3 completed
the SAS workload on
average 14 percent
faster than the
legacy server did.

Figure 3: The average time it
took each solution to complete
one SAS multi-user scenario in
hours:minutes:seconds.
Examining the time savings another way, the server powered by Intel Xeon
processors E7-8890 v3 completed an average of over 118 jobs in an hour, which is
nearly 13 times the number of jobs per hour that the legacy server performed. As shown
in Figure 4, the server powered by Intel Xeon processors E5-2680 v2 completed an
average of over 32 jobs per hour, or 2.77 times more jobs per hour than the legacy
server did. More SAS jobs per hour means your business can analyze more quickly.
Figure 4: The average number
of jobs per hour for each
solution.

In addition to the performance and time benefits, we saw that the four-socket
server powered by Intel Xeon processors E7-8890 v3 had system processing power
remaining even when it was running 12 virtual SAS 9.4 instances. This headroom means
that for some periods during the test scenario, the 12 VMs did not fully saturate the
processors, storage, or memory. With the available headroom, one could increase the
number of VMs on the server for more SAS 9.4 instances running additional analytics,
though we did not do so.
SUPPORTING PROCESSOR PERFORMANCE WITH INTEL SSD STORAGE
Running SAS 9.4 software on a server without improved storage performance
could place limitations on the workloads, leave the processor underutilized, and limit
the speed of your SAS jobs. Combining the Intel Xeon processors E5 v2 and E7 v3 with
the Intel DC S3700 SATA and DC P3700 PCIe SSDs respectively, allowed us to increase
the workload density and get more from our storage subsystem. A higher workload
density meant the newer servers completed more jobs simultaneously and averaged
more jobs per hour. Using the Intel DC S3700 SATA and DC P3700 PCIe SSDs improved
the outputs from the disk subsystem. Figure 5 shows the peak disk throughput that each
solution achieved.
Figure 5: The peak disk
throughput in GB per second
for each solution.
Even with our 12 VMs on the server powered by Intel Xeon processors E7-8890 v3 with Intel SSDs DC P3700,
we had headroom remaining at points in the test, which could translate to a potentially greater number of
VMs in your environment.

Your business could see a number of potential benefits by upgrading your legacy
servers with hard-disk drives (HDDs) to newer servers with SATA and PCIe SSDs:
 Get SAS data sooner and have more time to analyze the data
 Save on management time by having fewer servers to maintain
 Reduce waste by utilizing processor and memory resources more
effectively
CONCLUSION
A key to modernizing your data center is to consolidate your legacy workloads
through virtualization, which can help reduce complexity for your business. Fewer
servers require fewer physical resources, such as power, cabling, and switches, and
reduce the burden on IT for ongoing management tasks such as updates. In addition,
integrating newer hardware technology into your data center can provide new features
that strengthen your infrastructure, such as RAS features on the processor and disk
performance improvements. Finally, using SAS 9.4 ensures that you have the latest
features and toolsets that SAS can offer.
Compared to a legacy server, we found that a modern four-socket server
powered by Intel Xeon processors E7-8890 v3 with Intel SSD DC P3700 Series provided
12 times the amount of SAS work, nearly 14 times the relative performance, and a
shorter average time to complete the SAS workload. Running 12 virtual SAS instances
also left capacity on the server for additional work. Consolidating your SAS workloads
from legacy servers onto servers powered by Intel Xeon processors E7 v3 and SAS 9.4
can provide your business with the latest hardware and software features, reduce
complexity in your data center, and potentially reduce costs for your business.

APPENDIX A – ABOUT THE COMPONENTS
About the new Intel Xeon processor E7 v3 family
Intel designed the new Intel Xeon processor E7 v3 family to support mission-critical, high-performance
workloads by featuring up to 18 cores and 36 threads per socket to provide significant jumps in performance from
previous releases. The Intel Xeon processor E7 v3 family supports up to 1.5TB DDR4 memory per socket, supports up to
24 DDR4 DIMMs per socket, and supports up to 2,133MHz DDR4 speeds to improve performance and increase
scalability.
The Intel Xeon processor E7 v3 family supports all the previous reliability, availability, and serviceability features
of previous processor releases to support critical workloads. With Intel® Run Sure technology, these processors add new
RAS features, including eMCA Gen 1, MCA Recovery – Execution Path, MCA IO, and PCIe Live Error Recovery. For more
information about the Intel Xeon processor E7 v3 product family, visit www.intel.com.
About the new Intel SSD DC P3700 Series
According to Intel, “[t]he Intel® Solid-State Drive Data Center Family for PCIe brings extreme data throughput
directly to Intel Xeon processors with up to six times faster data transfer speed than 6 Gbps SAS/SATA SSDs. The
performance of a single drive from the Intel SSD Data Center Family for PCIe, specifically the Intel Solid-State Drive Data
Center P3700 Series (460K IOPS), can replace the performance of seven SATA SSDs aggregated through a host bus
adapter (HBA) (approximately 500K IOPS).” For more information on the Intel SSD PC P3700 Series, visit
www.intel.com/content/www/us/en/solid-state-drives/solid-state-drives-dc-p3700-series.html.
About SAS 9.4
SAS 9.4, the latest release of SAS®9 architecture, uses multicore technologies2
to deliver processing capabilities
through in-database and in-memory analytics. According to SAS, this results “in greater insights more quickly from big
data and streaming data.” SAS upgraded the architecture with features that meet the needs of traditional on-site SAS
deployments and of private and public cloud deployments:
 Choose from many deployment options, including hosted and managed cloud options with SAS Solutions
OnDemand
 Use potentially highly scalable environments for testing and development with six new products
 Integrate SAS into your business processes with new APIs, including mobile delivery options for popular
smartphones and tablets
 Get monitoring and management capabilities with the new Web-based SAS Environment Manager
 Monitor data according to your schedule (daily, weekly, or monthly) with the new SAS Environment
Manager
 Receive alerts, notifications, and data from your servers in the customizable SAS Environment Manager
dashboard
2
Multi-core/threading technologies were not fully leveraged with the workload used in this study, but can be with newer SAS 9.4
capabilities such as In-memory and high-performance analytics.

The second maintenance release, SAS® 9.4 TS1M2 (9.4M2), offers new features and enhancements for products
such as SAS/STAT®, SAS® Enterprise Miner™, and SAS/ETS®. For more information on SAS 9.4, visit
sas.com/en_us/software/sas9.html.
About the SAS workload
Test details
Each test scenario consisted of a set of 25 jobs running in a multiuser fashion to simulate a typical SAS batch and
SAS enterprise user environment where 8 to 16 simultaneous jobs were executing at any one time (except during initial
startup and during the ramp down period at the end). In the scenario-controlled scripts, jobs were launched over a set
interval to simulate a multiuser environment where users came and went from the system. To help simulate a typical
work environment during the test, some periods were heavier than others were. Some periods were batch-only and
other periods were batch and interactive users. This simulated a real-world environment where there are peaks and
valleys in system resource use.
The jobs used a mix of CPU, memory, and I/O-intensive workloads designed to simulate a larger work-group of 8
to 16 simultaneous sessions. During the test, 14 different programs were executed (25 different jobs were run during
the test, some more than once). We used different input data for all jobs. SAS procedures used during execution
included: LOGISTIC, REG, GLM, SORT, MEANS, MIXED, SUMMARY, DATA steps, and SQL.
Data
A single copy of the workload includes the following details:
• Individual file sizes up to 30 GB
• File types: text, sas7bdat, binary
• Row counts up to 76 million
• Variable and or column counts up to 297
• Binary file usage with EBCDIC and binary data from a mainframe (MVS)
• Approximately 130 GB of input data was used in each 8-to-16-simultaneous-user scenario that we executed
Input and output volumes were at least three to four times the input size (SAS programs typically read and write
data depending on function).

APPENDIX B – SYSTEM CONFIGURATION INFORMATION
Figure 6 provides detailed configuration information for the test systems.
System Dell PowerEdge R710 Dell PowerEdge R720 Dell PowerEdge R930
General
Number of processor
packages
2 2 4
Number of cores per
processor
4 10 18
Number of hardware
threads per core
2 2 2
System power
management policy
Maximum performance Performance Performance
Power supplies
Total number 2 2 4
Vendor and model
number
Dell A870P-00 Dell E1100E-S0 Dell E1100E-S0
Wattage of each (W) 870 1,100 1,100
Cooling fans
Total number 5 6 6
Vendor and model
number
Dell PFC0612DE
San Ace® 60
9GA0612P1K641
Nidec® UltraFlo®
V12C12BS1M3
Dimensions (h × w) of
each
3″ × 3″ 3″ × 3″ 6″ × 6″
Volts 12 12 12
Amps 1.68 0.95 2.31
CPU
Vendor Intel Intel Intel
Name Xeon Xeon Xeon
Model number X5570 E5-2680 v2 E7-8890 v3
Socket type LGA1366 LGA2011 LGA2011
Core frequency (GHz) 2.93 2.8 2.5
Bus frequency 6.4 GT/s 8 GT/s 9.6 GT/s
L1 cache
4 × 32 KB Data
4 × 32 KB Instr.
10 × 32 KB Data
10 × 32 KB Instr.
18 × 32 KB Data
18 × 32 KB Instr.
L2 cache 4 × 256 KB 10 × 256 KB 18 × 256 KB
L3 cache 8 MB 25 MB 45 MB
Platform
Vendor and model
number
Dell PowerEdge R710 Dell PowerEdge R720 Dell PowerEdge R930
Motherboard model
number
0YDJK3 0M1GCRX04 0Y4CNCA01
BIOS name and version 6.4.0 2.5.2 1.0.2 [MRC_096]
BIOS settings Default Default Default

Memory module(s)
Total RAM in system
(GB)
48 256 1,024
Vendor and model
number
Samsung® M393B1K70DH0-
YH9
Hynix HMT42GR7MFR4C-PB
Samsung M386A4G40DM0-
CPB
Type PC3L-10600R PC3-12800R PC4-17000R
Speed (MHz) 1,333 1,600 2,133
Speed running in the
system (MHz)
1,333 1,600 1,600
Timing/Latency (tCL-
tRCD-tRP-tRASmin)
9-9-9-36 11-11-11-35 15-15-15-36
Size (GB) 8 16 32
Number of RAM
module(s)
6 16 32
Chip organization Double Sided Double Sided Double Sided
Rank Dual Dual Quad
Operating system
Name CentOS 6.6 CentOS 6.6 CentOS 6.6
Kernel Version 2.6.32-504.16.2.el6.x86_64 2.6.32-504.16.2.el6.x86_64 2.6.32-504.16.2.el6.x86_64
File system
ext4 (OS and SAS program
files)
xfs (SAS DATA/OUTPUT)
xfs (SAS WORK/UTILLOC)
files)
files)
RAID controller
Vendor and model
number
Dell PERC 6/i Dell PERC H710P Mini Dell PERC H730P
Firmware version 6.3.3.0002 21.3.1-0004 25.2.2-0004
Driver version
06.803.01.00-rh1
(megaraid_sas)
06.803.01.00-rh1
(megaraid_sas)
06.803.01.00-rh1
(megaraid_sas)
Cache size (MB) 256 1,024 2,048
Solid-state drives
Vendor and model
number
N/A Intel SSD DC S3700 Intel SSD DC P3700
Number of drives N/A 8 8
Size N/A 800 GB 2.0 TB
Type N/A 2.5in SATA 6Gb/s 1/2 Height PCIe 3.0
Hard drives
Vendor and model
number
8 × Seagate® ST9300605SS
2 × Seagate ST300MM0006,
4 × Toshiba® MBF2600RC
2 × Toshiba AL13SEB900,
8 × Toshiba MBF2600RC
Number of drives 8 6 10
Size (GB) 300 300, 600 900, 600
RPM 15K 10K 10K
Type SAS 6Gb/s SAS 6Gb/s SAS 6Gb/s

Ethernet adapters
Vendor and model
number
Broadcom NetXtreme® II 4P Intel 2P X540/2P I350 Intel 2P X540/2P I350
Type rNDC rNDC rNDC
Driver bnx2 v2.2.4 Intel 5.0.5-k Intel 5.0.5-k
Figure 6: System configuration information for the test systems.

APPENDIX C – HOW WE TESTED
Installing the CentOS 6.6 64-bit operating system
1. Insert the CentOS 6.6 installation DVD, and boot from it.
2. On the Welcome to CentOS 6! screen, select Install or upgrade an existing system, and press Enter.
3. On the Disc Found screen, select Skip, and press Enter.
4. On the CentOS 6 screen, click Next.
5. On the installation-selection screen, keep the default, and click Next.
6. One the keyboard-selection screen, keep the default, and click Next.
7. On the storage-selection screen, click Basic Storage Devices, and click Next.
8. On the Storage Device Warning pop-up screen, click Yes, discard any data.
9. On the Hostname screen, enter the server’s name, and click Configure Network.
10. On the Network Connections pop-up screen, click Add.
11. On the Choose a Connection Type, select Wired, and click Create.
12. On the Editing Wired Connection pop-up, select the IPv4 Settings tab, change Method to Manual, click Add,
enter the interface's IP address, netmask, and gateway, and click Apply.
13. Close the Network Connections pop-up screen.
14. Click next on the Hostname screen.
15. On the time-zone screen, click Next.
16. On the administrator-password screen, enter the Root Password (twice), and click Next.
17. On the Which type of installation would you like screen, click both Replace Existing Linux Systems(s), and click
Next.
18. On the Format Warnings pop-up screen, click Format.
19. On the Writing storage configuration to disk pop-up screen, click Write changes to disk.
20. On the boot-loader selection screen, click Next.
21. On the software-selection screen, click Basic Server (R710, R720 VM, R930 VM) or Virtual Host (R720, R930), and
click Next.
22. On the Congratulations screen, click Reboot.
Configuring the Dell R710 (bare-metal)
Run the following commands for each process.
Installing updates and additional packages
yum install -y epel-release
yum update -y
yum install -y chrony nmon time xfsprogs tuned numactl acpid cpuspeed wget vim
nfs-utils openssh-clients man unzip smartmontools numactl ipmitool OpenIPMI
sysstat
Disabling SELINUX and the firewall
setenforce 0
sed -i 's/SELINUX=enforcing/SELINUX=disabled/' /etc/selinux/config
service iptables stop

service ip6tables stop
chkconfig iptables off
chkconfig ip6tables off
Editing /etc/fstab
Change the following:
/dev/mapper/vg_r710-lv_root / ext4 defaults 1 1
To the following:
/dev/mapper/vg_r710-lv_root / ext4 defaults,nobarrier,noatime,nodiratime 1 1
Synching the time
service ntpd stop
chkconfig ntpd off
sed -i '/server .*/d' /etc/chrony.conf
echo 'server time.ptnet.principledtech.com iburst prefer' >> /etc/chrony.conf
service chronyd restart
chkconfig chronyd on
Creating and formatting volumes (assume RAID10)
VG_NAME=vg_r710
DISKS=8
STRIPE=256
lvcreate ${VG_NAME} -L 350G -C y -n lv_sasdata
lvcreate ${VG_NAME} -L 250G -n lv_saswork
mkfs.xfs -d su=${STRIPE}k,sw=$(($DISKS / 2)) /dev/${VG_NAME}/lv_sasdata
mkfs.xfs -d su=${STRIPE}k,sw=$(($DISKS / 2)) /dev/${VG_NAME}/lv_saswork
mkdir -p /data
echo -e "/dev/${VG_NAME}/lv_sasdata /data xfs
defaults,nobarrier,noatime,nodiratime,inode64 0 0" >> /etc/fstab
mount -v /data
mkdir -p /data/saswork
echo "/dev/${VG_NAME}/lv_saswork /data/saswork xfs
mount -v /data/saswork
Creating and applying a tuned profile
cp -r /etc/tune-profiles/enterprise-storage /etc/tune-profiles/sas
sed -i 's/set_transparent_hugepages.*always/set_transparent_hugepages never/'
/etc/tune-profiles/sas/ktune.sh
sed -i 's/multiply_disk_readahead.*4/multiply_disk_readahead 64/' /etc/tune-
profiles/sas/ktune.sh
tuned-adm profile sas
reboot

Creating users
groupadd -g 500 sas
useradd -u 500 -g 500 sasdemo
useradd -u 400 -g 500 sas
passwd sasdemo
passwd sas
echo 'export PATH=$PATH:/opt/SAS/SASFoundation/9.4' >> /home/sasdemo/.bashrc
echo 'export ASUITE=/data/asuite' >> /home/sasdemo/.bashrc
echo 'export PATH=$PATH:/opt/SAS/SASFoundation/9.4' >> /root/.bashrc
echo 'export ASUITE=/data/asuite' >> /root/.bashrc
Extracting SAS workload files
cd /
tar -zxvf opt_SAS.tar.gz
tar -zxvf data.tar.gz
Configure the R720 (Virtual Host)
Run the following commands for each process.
yum update -y
sysstat
setenforce 0
Editing /etc/fstab
1. Change the following:
To the following:
2. Add the following:
hugetlbfs /dev/hugepages hugetlbfs mode=1770,gid=107 0 0
Editing /etc/libvirt/qemu.conf
hugetlbfs_mount = "/dev/hugepages"
Editing /etc/sysctl.conf
vm.hugetlb_shm_group = 107
vm.nr_hugepages = 122912

Synching the time
service ntpd stop
chkconfig ntpd off
STRIPE=256
VMCOUNT=4
POOL_NAME[0]=vmpool
POOL_DEVS[0]="/dev/sdb"
POOL_DISKS[0]=4
POOL_HWRAID[0]=true
POOL_VOLNAME[0]=os
POOL_VOLSIZE[0]=80
POOL_NAME[1]=datapool
POOL_DEVS[1]="/dev/sdc"
POOL_DISKS[1]=8
POOL_HWRAID[1]=true
POOL_VOLNAME[1]=data
POOL_VOLSIZE[1]=600
for (( i=0; i<${#POOL_NAME[@]}; i++ ));
do
NAME=${POOL_NAME[$i]}
DEVS=${POOL_DEVS[$i]}
DISKS=${POOL_DISKS[$i]}
HWRAID=${POOL_HWRAID[$i]}
VOLNAME=${POOL_VOLNAME[$i]}
VOLSIZE=${POOL_VOLSIZE[$i]}
vgremove -f $NAME
pvremove $DEVS
for dev in $DEVS; do
dd if=/dev/zero of=${dev} bs=1M count=10 oflag=direct
done
if $HWRAID; then
pvcreate --dataalignment $((${DISKS} / 2 * ${STRIPE}))K $DEVS
else
pvcreate $DEVS
fi
vgcreate -s 1g $NAME $DEVS
for vm in `seq 1 $VMCOUNT`;
do
if $HWRAID; then

lvcreate -n ${VOLNAME}${vm} -l ${VOLSIZE} -C y $NAME
else
if [ ${DISKS} -lt 4 ]; then
lvcreate --type raid1 -m 1 -n ${VOLNAME}${vm} -l ${VOLSIZE} -C y $NAME
else
lvcreate --type raid10 -i $((${DISKS} / 2)) -I ${STRIPE} -m 1 -n
${VOLNAME}${vm} -l ${VOLSIZE} -C y $NAME
fi
fi
done
virsh pool-define-as --name $NAME --type logical --target /dev/$NAME
virsh pool-start --pool $NAME
virsh pool-autostart --pool $NAME
sync
done
tuned-adm profile virtual-host
Configure the R930 (Virtual Host)
yum update -y
sysstat
setenforce 0
Editing /etc/fstab
1. Change the following:
To the following:
2. Add the following:
hugetlbfs /dev/hugepages hugetlbfs mode=1770,gid=107 0 0
Editing /etc/libvirt/qemu.conf
hugetlbfs_mount = "/dev/hugepages"

Editing /etc/sysctl.conf
vm.hugetlb_shm_group = 107
vm.nr_hugepages = 491616
Synching the time
service ntpd stop
chkconfig ntpd off
STRIPE=256
VMCOUNT=12
POOL_NAME[0]=vmpool
POOL_DEVS[0]="/dev/sdb"
POOL_DISKS[0]=4
POOL_HWRAID[0]=true
POOL_VOLNAME[0]=os
POOL_VOLSIZE[0]=80
for (( i=0; i<${#POOL_NAME[@]}; i++ ));
do
NAME=${POOL_NAME[$i]}
DEVS=${POOL_DEVS[$i]}
DISKS=${POOL_DISKS[$i]}
HWRAID=${POOL_HWRAID[$i]}
VOLNAME=${POOL_VOLNAME[$i]}
VOLSIZE=${POOL_VOLSIZE[$i]}
vgremove -f $NAME
pvremove $DEVS
for dev in $DEVS; do
dd if=/dev/zero of=${dev} bs=1M count=10 oflag=direct
done
if $HWRAID; then
pvcreate --dataalignment $((${DISKS} / 2 * ${STRIPE}))K $DEVS
else
pvcreate $DEVS
fi
vgcreate -s 1g $NAME $DEVS
for vm in `seq 1 $VMCOUNT`;
do
if $HWRAID; then
lvcreate -n ${VOLNAME}${vm} -l ${VOLSIZE} -C y $NAME
else
if [ ${DISKS} -lt 4 ]; then

lvcreate --type raid1 -m 1 -n ${VOLNAME}${vm} -l ${VOLSIZE} -C y $NAME
else
lvcreate --type raid10 -i $((${DISKS} / 2)) -I ${STRIPE} -m 1 -n
${VOLNAME}${vm} -l ${VOLSIZE} -C y $NAME
fi
fi
done
virsh pool-define-as --name $NAME --type logical --target /dev/$NAME
virsh pool-start --pool $NAME
virsh pool-autostart --pool $NAME
sync
done
Creating and formatting volumes (using LVM RAID1)
vgcreate -s 1g datapool /dev/nvme0n1 /dev/nvme1n1 /dev/nvme2n1 /dev/nvme3n1
/dev/nvme4n1 /dev/nvme5n1 /dev/nvme6n1 /dev/nvme7n1
lvcreate --type raid1 -m 1 -n data1 -l 600 -C y datapool /dev/nvme0n1
/dev/nvme1n1
/dev/nvme3n1
/dev/nvme5n1
/dev/nvme7n1
/dev/nvme0n1
/dev/nvme2n1
/dev/nvme4n1
/dev/nvme6n1
/dev/nvme1n1
/dev/nvme3n1
/dev/nvme5n1
/dev/nvme7n1
Configure the R720 Virtual Machines
yum update -y
yum install -y chrony nmon time xfsprogs tuned acpid wget vim nfs-utils openssh-
clients man unzip sysstat numactl

ln -s /dev/null /etc/udev/rules.d/75-persistent-net-generator.rules
rm -f /etc/udev/rules.d/70-persistent-net.rules
setenforce 0
Editing /etc/fstab
/dev/mapper/vg_r720vm1-lv_root / ext4 defaults 1 1
To the following:
/dev/mapper/vg_r720vm1-lv_root / ext4 defaults,nobarrier,noatime,nodiratime 1 1
Synching the time
service ntpd stop
chkconfig ntpd off
Creating users
groupadd -g 500 sas
passwd sasdemo
passwd sas
reboot
VG_NAME=vg_sas
STRIPE=256

DATA_DISKS=8
WORK_DISKS=8
DATA_DEV=/dev/vdb
WORK_DEV=/dev/vdb
pvcreate --dataalignment $((${DATA_DISKS} / 2 * ${STRIPE}))K $DATA_DEV
pvcreate --dataalignment $((${WORK_DISKS} / 2 * ${STRIPE}))K $WORK_DEV
vgcreate -s ${STRIPE}m $VG_NAME $DATA_DEV
vgextend $VG_NAME $WORK_DEV
lvcreate ${VG_NAME} ${DATA_DEV} -L 350G -n lv_sasdata
lvcreate ${VG_NAME} ${WORK_DEV} -l 100%FREE -n lv_saswork
mkfs.xfs -d su=${STRIPE}k,sw=$((${DATA_DISKS} / 2)) /dev/${VG_NAME}/lv_sasdata
mkfs.xfs -d su=${STRIPE}k,sw=$((${WORK_DISKS} / 2)) /dev/${VG_NAME}/lv_saswork
mkdir -p /data
mount -v /data
poweroff
cd / ; tar -zxvf opt_SAS.tar.gz
cd / ; tar -zxvf data.tar.gz
Configure the R920 Virtual Machines
yum update -y
yum install -y chrony nmon time xfsprogs tuned acpid wget vim nfs-utils openssh-
clients man unzip sysstat numactl
ln -s /dev/null /etc/udev/rules.d/75-persistent-net-generator.rules
rm -f /etc/udev/rules.d/70-persistent-net.rules
setenforce 0

Editing /etc/fstab
/dev/mapper/vg_r920vm1-lv_root / ext4 defaults 1 1
To the following:
/dev/mapper/vg_r920vm1-lv_root / ext4 defaults,nobarrier,noatime,nodiratime 1 1
Synching the time
service ntpd stop
chkconfig ntpd off
Creating users
groupadd -g 500 sas
passwd sasdemo
passwd sas
reboot
Creating and formatting volumes (assume no RAID)
VG_NAME=vg_sas
STRIPE=256
DATA_DISKS=2
WORK_DISKS=2
DATA_DEV=/dev/vdb
WORK_DEV=/dev/vdb
pvcreate $DATA_DEV
pvcreate $WORK_DEV
vgcreate -s ${STRIPE}m $VG_NAME $DATA_DEV
vgextend $VG_NAME $WORK_DEV

lvcreate ${VG_NAME} ${DATA_DEV} -L 350G -n lv_sasdata
lvcreate ${VG_NAME} ${WORK_DEV} -l 100%FREE -n lv_saswork
mkfs.xfs -d su=${STRIPE}k,sw=${DATA_DISKS} /dev/${VG_NAME}/lv_sasdata
mkfs.xfs -d su=${STRIPE}k,sw=${WORK_DISKS} /dev/${VG_NAME}/lv_saswork
mkdir -p /data
mount -v /data
poweroff
cd / ; tar -zxvf opt_SAS.tar.gz
cd / ; tar -zxvf data.tar.gz

APPENDIX D – CONFIGURATION FILES
R720-vm1.xml
<domain type='kvm' id='1'>
<name>r720-vm1</name>
<uuid>95e5115b-fe41-9a1a-4006-26d90eed832d</uuid>
<memory unit='KiB'> 62914560</memory>
<currentMemory unit='KiB'> 62914560</currentMemory>
<memoryBacking>
<hugepages/>
<nosharepages/>
</memoryBacking>
<vcpu placement='static'
cpuset='0,2,4,6,8,10,12,14,16,18,20,22,24,26,28,30,32,34,36,38'>8</vcpu>
<numatune>
<memory mode='strict' nodeset='0'/>
</numatune>
<os>
<type arch='x86_64' machine='rhel6.6.0'>hvm</type>
<boot dev='hd'/>
</os>
<features>
<acpi/>
<apic/>
<pae/>
</features>
<cpu mode='host-passthrough'>
</cpu>
<clock offset='utc'/>
<on_poweroff>destroy</on_poweroff>
<on_reboot>restart</on_reboot>
<on_crash>restart</on_crash>
<devices>
<emulator>/usr/libexec/qemu-kvm</emulator>
<disk type='block' device='disk'>
<driver name='qemu' type='raw' cache='none' io='native'/>
<source dev='/dev/vmpool/os1'/>
<target dev='vda' bus='virtio'/>
<alias name='virtio-disk0'/>
<address type='pci' domain='0x0000' bus='0x00' slot='0x05' function='0x0'/>
</disk>
<source dev='/dev/datapool/data1'/>
<target dev='vdb' bus='virtio'/>
<alias name='virtio-disk1'/>
</disk>
<disk type='block' device='cdrom'>
<driver name='qemu' type='raw'/>
<target dev='hdc' bus='ide'/>

<listen type='address' address='127.0.0.1'/>
</graphics>
<video>
<model type='cirrus' vram='9216' heads='1'/>
<alias name='video0'/>
</video>
<memballoon model='virtio'>
<alias name='balloon0'/>
</memballoon>
</devices>
</domain>
R920-vm1.xml
<domain type='kvm'>
<name>r930-vm1</name>
<uuid>9e770a15-6fa8-41ed-466d-c3a31f3aaada</uuid>
<memory unit='KiB'>83886080</memory>
<currentMemory unit='KiB'>83886080</currentMemory>
<memoryBacking>
<hugepages/>
<nosharepages/>
</memoryBacking>
<vcpu placement='static'
cpuset='0,4,8,12,16,20,24,28,32,36,40,44,48,52,56,60,64,68,72,76,80,84,88,92,96,1
00,104,108,112,116,120,124,128,132,136,140'>8</vcpu>
<numatune>
<memory mode='strict' nodeset='0'/>
</numatune>
<os>
<type arch='x86_64' machine='rhel6.6.0'>hvm</type>
<boot dev='hd'/>
</os>
<features>
<acpi/>
<apic/>
<pae/>
</features>
<cpu mode='host-passthrough'>
</cpu>
<clock offset='utc'/>
<on_poweroff>destroy</on_poweroff>
<on_reboot>restart</on_reboot>
<on_crash>restart</on_crash>
<devices>
<emulator>/usr/libexec/qemu-kvm</emulator>
<source dev='/dev/vmpool/os1'/>
<target dev='vda' bus='virtio'/>

</disk>
<source dev='/dev/datapool/data1'/>
<target dev='vdb' bus='virtio'/>
</disk>
<disk type='block' device='cdrom'>
<driver name='qemu' type='raw'/>
<target dev='hdc' bus='ide'/>
<readonly/>
<address type='drive' controller='0' bus='1' target='0' unit='0'/>
</disk>
<controller type='usb' index='0' model='ich9-ehci1'>
</controller>
<address type='pci' domain='0x0000' bus='0x00' slot='0x04' function='0x0'
multifunction='on'/>
</controller>
</controller>
</controller>
<controller type='ide' index='0'>
</controller>
<interface type='bridge'>
<mac address='52:54:00:41:34:a3'/>
<source bridge='br0'/>
<model type='virtio'/>
</interface>
<serial type='pty'>
<target port='0'/>
</serial>
<console type='pty'>
<target type='serial' port='0'/>
</console>
<input type='tablet' bus='usb'/>
<input type='mouse' bus='ps2'/>
<graphics type='vnc' port='-1' autoport='yes' keymap='en-us'/>
<video>
<model type='cirrus' vram='9216' heads='1'/>
</video>

APPENDIX E – TEST SCRIPTS
run_sas_multi.sh
#!/bin/bash
STATS_INTERVAL=15
MAX_RUNTIME=7200
REMOTE_ASUITE=/data/asuite
REMOTE_SASPATH=/opt/SAS/SASFoundation/9.4
SSH_CMD="ssh -2"
MAX_SAMPLES=$((${MAX_RUNTIME} / ${STATS_INTERVAL}))
TIMESTAMP=$(date +"%Y%m%d_%H%M%S")
if [ "$SAS_HOSTS" == "" ]; then
echo Variable SAS_HOSTS is invalid...exiting
exit
fi
echo "SAS Hosts: $SAS_HOSTS"
# Check if VM
VM_TEST=true
if [ "$VIRT_HOSTS" == "$SAS_HOSTS" ]; then
VM_TEST=false
elif [ "$VIRT_HOSTS" == "" ]; then
VM_TEST=false
VIRT_HOSTS="$SAS_HOSTS"
fi
if $VM_TEST; then
echo "Hypervisor Hosts: $VIRT_HOSTS"
fi
TEST_NAME=$(echo $VIRT_HOSTS | awk '{print $1}')
RESULTS_DIR="${TEST_NAME}_${TIMESTAMP}"
REMOTE_HOST=
echo "Results Directory: $RESULTS_DIR"
mkdir -p $RESULTS_DIR
# Cleanup and drop caches
echo Cleanup and drop caches
for REMOTE_HOST in $SAS_HOSTS;
do
$SSH_CMD root@$REMOTE_HOST "export ASUITE=${REMOTE_ASUITE} ; pkill sas ; sync ;
sleep 1 ; $ASUITE/bin/clean.sh ; rm -f $ASUITE/logs/*.* ; sync ; sleep 1 ; echo
3 > /proc/sys/vm/drop_caches ; sync ; sleep 1 ; find /* > /dev/null" 1> /dev/null
&
done
wait

# Start performance monitoring on VIRT hosts
if $VM_TEST; then
echo Start performance monitoring on VIRT hosts
for REMOTE_HOST in $VIRT_HOSTS;
do
$SSH_CMD root@$REMOTE_HOST "pkill nmon ; rm -f /tmp/*.nmon ; nmon -F
/tmp/${REMOTE_HOST}.nmon -s${STATS_INTERVAL} -c${MAX_SAMPLES} -t -C sas"
$SSH_CMD -f root@$REMOTE_HOST "pkill sar ; rm -f /tmp/*.sar ; sar -o
/tmp/${REMOTE_HOST}.sar ${STATS_INTERVAL} ${MAX_SAMPLES} > /dev/null 2>&1" 2>
/dev/null
done
sleep $((${STATS_INTERVAL} * 2))
fi
# Start performance monitoring on SAS hosts
echo Start performance monitoring on SAS hosts
do
$SSH_CMD root@$REMOTE_HOST "pkill nmon ; nmon -F
${REMOTE_ASUITE}/logs/${REMOTE_HOST}.nmon -s${STATS_INTERVAL} -c${MAX_SAMPLES} -t
-C sas"
$SSH_CMD -f root@$REMOTE_HOST "pkill sar ; sar -o
${REMOTE_ASUITE}/logs/${REMOTE_HOST}.sar ${STATS_INTERVAL} ${MAX_SAMPLES} >
/dev/null 2>&1" 2> /dev/null
done
# Run SAS Demo
date +%s.%N > ${RESULTS_DIR}/start_time.txt
do
#/usr/bin/time -p $SSH_CMD sasdemo@$REMOTE_HOST "export ASUITE=${REMOTE_ASUITE}
; export PATH=$PATH:$REMOTE_SASPATH ; echo Running on ${REMOTE_HOST}!
PATH=$PATH 2> $ASUITE/logs/time.log | tee $ASUITE/logs/output.log ; sleep 5 ;
echo Done on ${REMOTE_HOST}!" 2> ${RESULTS_DIR}/${REMOTE_HOST}_time.log 1>
${RESULTS_DIR}/${REMOTE_HOST}_output.log &
/usr/bin/time -p $SSH_CMD sasdemo@$REMOTE_HOST "/usr/bin/time -p
$ASUITE/bin/mixed_8_to_16.sh 2> $ASUITE/logs/time.log | tee
$ASUITE/logs/output.log ; sync" 2> ${RESULTS_DIR}/${REMOTE_HOST}_time.log 1>
${RESULTS_DIR}/${REMOTE_HOST}_output.log &
done
sleep 1
echo "Running SAS Demo! (To see output run: tail -f ${RESULTS_DIR}/*_output.log)"
wait
date +%s.%N > ${RESULTS_DIR}/stop_time.txt
echo SAS Demo complete!
sync
echo "$(cat ${RESULTS_DIR}/stop_time.txt) - $(cat ${RESULTS_DIR}/start_time.txt)"
| bc > ${RESULTS_DIR}/test_time.txt

# Stop performance monitoring on SAS hosts
echo Stop performance monitoring on SAS hosts
do
$SSH_CMD root@$REMOTE_HOST "pkill -USR2 nmon ; killall -w sar ; sync"
done
# Stop performance monitoring on VIRT hosts
if $VM_TEST; then
echo Stop performance monitoring on VIRT hosts
do
$SSH_CMD root@$REMOTE_HOST "pkill -USR2 nmon ; killall -w sar ; sync"
done
fi
# Parse and transfer remote results
echo Parse and transfer remote results
do
$SSH_CMD root@$REMOTE_HOST "export ASUITE=${REMOTE_ASUITE} ; cd $ASUITE/logs ;
grep -A100 "NOTE: The SAS System used:" *.log | grep -v "The SAS System" |
sed 's/.log-//' > sas_system.log ; cat sas_system.log | awk '/real time/{print
$1","$4}' | awk -F'[,:]'
'BEGIN{min_sum=0;sec_sum=0};/:/{printf("%s,%sn",$1,($2*60)+$3);min_sum+=$2
;sec_sum+=$3};!/:/{printf("%s,%sn",$1,$2);sec_sum+=$2};END{printf("total_
real_time,%sn",min_sum*60+sec_sum)}' > real_time.csv"
mkdir -p ${RESULTS_DIR}/${REMOTE_HOST}
scp root@$REMOTE_HOST:${REMOTE_ASUITE}/logs/*.* ${RESULTS_DIR}/${REMOTE_HOST}
done
OLD_DIR=$(pwd)
cd ${RESULTS_DIR}
find */real_time.csv | awk -F'/' '{printf("jobname,%s,",$1)}END{print""}' >
real_time.csv
paste -d',' */real_time.csv >> real_time.csv
cd ${OLD_DIR}
# Transfer virthost stats
if $VM_TEST; then
echo Transfer virthost stats
do
scp root@$REMOTE_HOST:/tmp/*.nmon ${RESULTS_DIR}/
scp root@$REMOTE_HOST:/tmp/*.sar ${RESULTS_DIR}/
for SAS_HOST in $SAS_HOSTS;
do
$SSH_CMD root@$REMOTE_HOST "virsh dumpxml ${SAS_HOST}" >
${RESULTS_DIR}/${SAS_HOST}.xml
done

done
fi
mixed_8_to_16.sh
#!/bin/sh
echo "computation multi-user ~8-16 simultaneous session test..."
echo ""
cd $ASUITE
OPTIONS="-memsize 2048M -bufsize 256k -sortsize 256M -fullstimer -work
/data/saswork "
OPTIONS2="-memsize 4096M -bufsize 256k -sortsize 256M -fullstimer -work
/data/saswork "
date
echo "ramp up..."
sas $ASUITE/code/codegen_issue.sas -sysparm 1 -log
$ASUITE/logs/codegen_a_1.log -print $ASUITE/lst/codegen_a_1.lst
$OPTIONS &
sleep 30
sas $ASUITE/code/io1_ca.sas -sysparm ca1 -log $ASUITE/logs/io1_ca_a_1.log
-print $ASUITE/lst/io1_ca_a_1.lst $OPTIONS &
sleep 30
sas $ASUITE/code/comp_test1.sas -log $ASUITE/logs/comp_test1_a_1.log
-print $ASUITE/lst/comp_test1_a_1.lst $OPTIONS &
sleep 30
sas $ASUITE/code/mixed4.sas -sysparm 1 -log $ASUITE/logs/mixed4_a_1.log
-print $ASUITE/lst/mixed4_a_1.lst $OPTIONS &
sleep 30
sas $ASUITE/code/dim.sas -sysparm 1 -log $ASUITE/logs/dim_a_1.log -print
$ASUITE/lst/dim_a_1.lst $OPTIONS &
sleep 30
sas $ASUITE/code/hist_clm.sas -sysparm 1 -log $ASUITE/logs/hist_clm_a_1.log
-print $ASUITE/lst/hist_clm_a_1.lst $OPTIONS &
sleep 30
sleep 90
sas $ASUITE/code/customer1.sas -sysparm 1 -log $ASUITE/logs/customer1_a_1.log
-print $ASUITE/lst/customer1_a_1.lst $OPTIONS &
sleep 5

sas $ASUITE/code/gsort.sas -sysparm 1 -log $ASUITE/logs/gsort_a_1.log
-print $ASUITE/lst/gsort_a_1.lst $OPTIONS &
sleep 60
sleep 30
sas $ASUITE/code/customer1.sas -sysparm 2 -log
$ASUITE/logs/customer1_a_2.log -print $ASUITE/lst/customer1_a_2.lst
$OPTIONS &
sleep 30
sas $ASUITE/code/io2.sas -sysparm large1 -log $ASUITE/logs/io2_a_1.log
-print $ASUITE/lst/io2_a_1.lst $OPTIONS &
sleep 30
sas $ASUITE/code/ranrw.sas -sysparm 1 -log $ASUITE/logs/ranrw_a_1.log
-print $ASUITE/lst/ranrw_a_1.lst $OPTIONS &
sleep 60
sleep 60
sleep 60
sas $ASUITE/code/ds.sas -sysparm 1 -log $ASUITE/logs/ds_a_1.log
-print $ASUITE/lst/ds_a_1.lst $OPTIONS &
sleep 120
sas $ASUITE/code/ds.sas -sysparm 2 -log $ASUITE/logs/ds_a_2.log
-print $ASUITE/lst/ds_a_2.lst $OPTIONS &
sleep 5
echo "All jobs are submitted!!!! "
wait
echo "All Done!!!!"
date
echo ""
echo "************************************************************************"
echo "*** ERROR CHECK - if there are ERRORS between the stars!! Check run!!***"

echo "************************************************************************"
echo "ERRORS: "
grep "ERROR:" $ASUITE/logs/*.log
echo "************************************************************************"
echo "************************************************************************"
echo ""

ABOUT PRINCIPLED TECHNOLOGIES
Principled Technologies, Inc.
1007 Slater Road, Suite 300
Durham, NC, 27703
www.principledtechnologies.com
We provide industry-leading technology assessment and fact-based
marketing services. We bring to every assignment extensive experience
with and expertise in all aspects of technology testing and analysis, from
researching new technologies, to developing new methodologies, to
testing with existing and new tools.
When the assessment is complete, we know how to present the results to
a broad range of target audiences. We provide our clients with the
materials they need, from market-focused data to use in their own
collateral to custom sales aids, such as test reports, performance
assessments, and white papers. Every document reflects the results of
our trusted independent analysis.
We provide customized services that focus on our clients’ individual
requirements. Whether the technology involves hardware, software, Web
sites, or services, we offer the experience, expertise, and tools to help our
clients assess how it will fare against its competition, its performance, its
market readiness, and its quality and reliability.
Our founders, Mark L. Van Name and Bill Catchings, have worked
together in technology assessment for over 20 years. As journalists, they
published over a thousand articles on a wide array of technology subjects.
They created and led the Ziff-Davis Benchmark Operation, which
developed such industry-standard benchmarks as Ziff Davis Media’s
Winstone and WebBench. They founded and led eTesting Labs, and after
the acquisition of that company by Lionbridge Technologies were the
head and CTO of VeriTest.
Principled Technologies is a registered trademark of Principled Technologies, Inc.
All other product names are the trademarks of their respective owners.
Disclaimer of Warranties; Limitation of Liability:
PRINCIPLED TECHNOLOGIES, INC. HAS MADE REASONABLE EFFORTS TO ENSURE THE ACCURACY AND VALIDITY OF ITS TESTING, HOWEVER,
PRINCIPLED TECHNOLOGIES, INC. SPECIFICALLY DISCLAIMS ANY WARRANTY, EXPRESSED OR IMPLIED, RELATING TO THE TEST RESULTS AND
ANALYSIS, THEIR ACCURACY, COMPLETENESS OR QUALITY, INCLUDING ANY IMPLIED WARRANTY OF FITNESS FOR ANY PARTICULAR PURPOSE.
ALL PERSONS OR ENTITIES RELYING ON THE RESULTS OF ANY TESTING DO SO AT THEIR OWN RISK, AND AGREE THAT PRINCIPLED
TECHNOLOGIES, INC., ITS EMPLOYEES AND ITS SUBCONTRACTORS SHALL HAVE NO LIABILITY WHATSOEVER FROM ANY CLAIM OF LOSS OR
DAMAGE ON ACCOUNT OF ANY ALLEGED ERROR OR DEFECT IN ANY TESTING PROCEDURE OR RESULT.
IN NO EVENT SHALL PRINCIPLED TECHNOLOGIES, INC. BE LIABLE FOR INDIRECT, SPECIAL, INCIDENTAL, OR CONSEQUENTIAL DAMAGES IN
CONNECTION WITH ITS TESTING, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. IN NO EVENT SHALL PRINCIPLED TECHNOLOGIES,
INC.’S LIABILITY, INCLUDING FOR DIRECT DAMAGES, EXCEED THE AMOUNTS PAID IN CONNECTION WITH PRINCIPLED TECHNOLOGIES, INC.’S
TESTING. CUSTOMER’S SOLE AND EXCLUSIVE REMEDIES ARE AS SET FORTH HEREIN.

Consolidate SAS 9.4 workloads with Intel Xeon processor E7 v3 and Intel SSD technology

Recommended

Recommended

More Related Content

What's hot

What's hot (20)

Similar to Consolidate SAS 9.4 workloads with Intel Xeon processor E7 v3 and Intel SSD technology

Similar to Consolidate SAS 9.4 workloads with Intel Xeon processor E7 v3 and Intel SSD technology (20)

More from Principled Technologies

More from Principled Technologies (20)

Recently uploaded

Recently uploaded (20)

Consolidate SAS 9.4 workloads with Intel Xeon processor E7 v3 and Intel SSD technology